Spiking Neural Networks, deep (reinforcement) learning in the brain, neural decision-making, continuous-time neural processing.
Amsterdam, The Netherlands
Joined July 2016
- Tweets52
- Following54
- Followers250
- Likes74
As a reviewer for ICRL, going over the other reviews posted I see 5/16 of those reviews (for 4 papers) clearly being generated by LLMs. Interesting strategy: vacuous comments and then scoring the paper quite high -- authors sure wont complain.
The yearly Neuromorphic Computing Netherlands meeting is coming up, Sept 19th, at TU Delft. With talks from Mark van Rossum, Sumeet Kumar, Wilfred van de Wiel and Mario Negrello, many posters and many friends. Official site: sites.google.com/view/ncn202…
BohteNeuroLabCWI retweeted
My esteemed @Cambridge_Uni colleague @lawrennd on the actually-existing threat from Big AI: archive.ph/2023.06.11-065606…
A great place to work and fascinating topic.
This Post is from an account that no longer exists. Learn more
Fun and exciting to talk on national television about spiking neural networks and energy efficient AI modeled after the brain. Dating myself: long ways from an obscure scientific topic (SNNs) in a highly unpopular machine learning domain (ANNs).
.@SanderBohte was interviewed by @Nieuwsuur on the enormous amount of power needed by #ArtificialIntelligence. He spoke about his research to capture the energy efficiency of our brains in an #algorithm and embed it on a chip. Watch the item here bit.ly/3INOlQY
Beyond reading the pretty pictures in this awesome reporting by @anilananth,
amazing to being mentioned in "Nature"...
AI models are growing in complexity at an impressive rate. But is a larger model necessarily smarter? A group of scientists is hoping to trim some of the fat, @anilananth reports for @Nature.
nature.com/articles/d41586-0…
Mooi stuk van Theo van Vugt, was een goed gesprek over de impact van foundational AI models zoals ChatGPT.
ChatGPT heeft de wereld veroverd. Het is kunstmatige intelligentie in werkende staat en voor het eerst wordt duidelijk wat de implicaties zijn voor gebruikers en de samenleving. De mogelijkheden zijn oneindig, maar gevaren zijn er ook.buff.ly/3ZNLONv
Organizing together w/ Sebastian Otte:
"Special Session on Recent Advances in Spiking Neural Networks" at ICANN 2023. Conference 26 - 29 Sep 2023, in beautiful Crete. Paper deadline mid-April.
#icann #enns #spiking #neuromorphic
e-nns.org/icann2023/wp-conte…
BohteNeuroLabCWI retweeted
Are you a PhD student with an interest in #MachineLearning? Sign up for our ML theory boot camp on the 14th and 15th of February: bit.ly/3Xz3Gd7
Bojian Yin demonstrating an SNN on a phone for keywordspotting at Science Park open dag (CWI), till 1700h today.
Great meeting yesterday in Nijmegen, wonderful to meet everybody in person again, and also to see so many new faces. Doubled from last year, neuromorphics rapidly becomings a big field in NL!
@PLanillos is opening today's symposium on Neuromorphic Computation in the Netherlands NCN2022 at the @DondersInst
Een eer en een genoegen om in Paradiso te staan. Bucket-list--
On Sunday, Sander Bohté (@CWInl, @UvA_Amsterdam, @univgroningen) took us on a fascinating journey in his #ParadisoLecture on the 'soul of AI': from the working of the brain (hearing neurons firing seeing a vertical line) to the basics of AI, his own work and future challenges.
Ik sta 19 juni in Paradiso om te vertellen over AI en het brein.
Info en kaartjes: paradiso.nl/nl/programma/par…
(lecture will be in Dutch)
Too modest: Lynn did some amazing work to develop recurrent models of object-recognition in the brain and map them to real human behavior.
💫Preprint time!💫
I’m excited to share our latest preprint “Mechanisms of human dynamic object recognition revealed by sequential deep neural networks” together with @SanderBohte, @dorinadejong, @HeleenSlagter & @hsteven9!
🧵 👇
Koen Vervloesem schrijft over energie-zuinige AI chips op Tweakers.net (registratie nodig voor volledig artikel): tweakers.net/reviews/9916/de…
BohteNeuroLabCWI retweeted
We might soon see neural networks that sip rather than gulp energy. My story on spiking #neuralnetworks for @SimonsInstitute, featuring work by @SanderBohte, @hisspikeness (Friedemann Zenke), @neuralreckoning (Dan Goodman) and @ChristianPehle : simons.berkeley.edu/news/spi…
BohteNeuroLabCWI retweeted
Recordings from the SNUFA SNN workshop are now available on YouTube:
youtube.com/playlist?list=PL…
cc @hisspikeness @SanderBohte
Very proud of this big effort, making artificial spiking neural networks performant and demonstrating energy efficiency.
New paper: effective and efficient spiking neural networks, closing the performance gap with classical recurrent neural networks while being vastly more energy efficient. Joint work with @SanderBohte (CWI) and Federico Corradi (IMEC).
Link (paywall):nature.com/articles/s42256-0…
Second edition: this should again be an on-point affair.
We'll hold the 2nd "SNUFA" spiking neural network online workshop this Nov 8-9. Free reg and abstract submission now open at:
snufa.net/2021
Invited speakers Tara Hamilton, @DylanRMuir, @cdschuman, and @sprekeler.
Posters, contributed talks and debate TBA
Please RT
